Machine Learning: Challenges and Opportunities Ahead |
Foundation of Computer Science USA |
MLCONF - Number 1 |
April 2014 |
Authors: S.yagasudha, L.latha, S.thangasamy |
6caa9f62-73c7-444f-b02f-ab97e7472367 |
S.yagasudha, L.latha, S.thangasamy . User Verification by Hierarchical Palmprint Matching. Machine Learning: Challenges and Opportunities Ahead. MLCONF, 1 (April 2014), 23-26.
Biometric verification system has high efficiency, high recognition rate and comfortable to user's operating characteristics. Palmprint authentication system is considered to be the most reliable biometric recognition technique due to its merits such as low-cost, user-friendliness, high speed and accuracy. Real time images are captured using a scanner at a resolution of 550 x 460. Each of these gray-scale images are aligned and then used to extract palmprint features. These features are then used for authenticating users. This paper presents a hierarchical palmprint matching system that is used to reduce the computation cost by segmenting the image and matching it with the database, thereby false palmprints are rejected in the subsequent changes by comparing just a portion of the whole palmprint.